INSIGHTS FROM THE ORIGINAL DATA

SAMPLE COMPOSITION

The Labour Force Survey sample can be thought of as comprising eight sub-samples (or rotation groups), with each sub-sample remaining in the survey for eight months, and one rotation group "rotating out" each month and being replaced by a new group "rotating in". This sample rotation is important in ensuring that seven-eighths of the sample are common from one month to the next, to ensure that changes in the estimates reflect real changes in the labour market, rather than the sample. In addition, the replacement sample is generally selected from the same geographic areas as the outgoing one, as part of a representative sampling approach.

When considering movements in the original estimates, it is possible to decompose the sample into three components:

• the matched common sample (survey respondents who responded in both February and March);
• the unmatched common sample (survey respondents who responded in March but who did not respond in February, or vice versa); and
• the incoming rotation group (survey respondents who replaced respondents who rotated out in February).

The detailed decomposition of each of these movements is included in the data cube 'Insights From the Original Data'.

In considering the three components of the sample, it is important to remember that the matched common sample describes the change observed for the same respondents in February and March, while the other two components reflect differences between the aggregate labour force status of different groups of people.

While the rotation groups are designed to be representative of the population, the outgoing and incoming rotation groups will almost always have somewhat different characteristics, as a result of the groups representing a sample of different households and people. The design of the survey, including the weighting and estimation processes, ensures that these differences are generally relatively minor and seeks to ensure that differences in characteristics of rotation groups do not affect the representativeness of the survey and its estimates. Monthly estimates are always designed to be representative of their respective months, regardless of the relative contribution of the three components of the sample.


MANAGING THE IMPACT OF THE TOWNSVILLE FLOODS ON LABOUR FORCE DATA

Flooding in Townsville in February 2019 resulted in a major disruption to the operation of the Labour Force Survey. As a result, there was a very low sample of responding households in the region in February.

Given the severity of these disruptions, which affected almost the entire region, and to ensure that this loss of sample did not affect data for Australia and Queensland, the ABS imputed sample for Townsville for February 2019. The imputation drew upon previous information that had recently been collected from people in Townsville.

Following the collection of March 2019 data from Townsville, the February data was re-assessed and has not been revised. The ABS will continue to monitor the data over coming months and in February 2020.

The imputation may have resulted in a slight overestimation of hours worked in Queensland in February 2019, given hours worked in Townsville will have been lower during the floods.

Users of the matched sample analysis should also exercise some caution when looking at Queensland data between January, February and March, given the effect of the imputation for Townsville.

During 2018, the ABS estimated that employed persons in Townsville accounted for around 1 per cent of all employed persons in Australia, around 4 per cent of employed persons in Queensland, and around 9 per cent of employed persons in the regions in Queensland outside of Brisbane.


INCOMING ROTATION GROUP

In original terms, the incoming rotation group in March 2019 had a higher employment to population ratio than the group it replaced (62.9% in February, compared to 63.9% in March 2019), and was higher than the ratio for the entire sample (62.4%). The full-time employment to population ratio of the incoming rotation group was higher than the group it replaced (43.9% in February, compared to 44.8% in March 2019), and was higher than the ratio for the entire sample (42.8%).

The unemployment rate of the incoming rotation group was 0.6 pts higher than the group it replaced (5.1% in February, compared to 5.7% in March 2019) and was higher that the unemployment rate for the whole sample (5.4%). The participation rate was 1.5 pts higher than the group it replaced (66.3% in February, compared to 67.7% in March 2019) and was higher than the sample as a whole (65.9%).


OUTGOING ROTATION GROUP

In looking ahead to the April 2019 estimates, in original terms, the outgoing rotation group in March 2019, which will be replaced by a new incoming rotation group in April 2019, has a higher employment to population ratio (62.6% in March 2019) compared to the sample as a whole (62.4%). The full-time employment to population ratio (42.5%) is lower than the ratio for the entire sample (42.8%).

The unemployment rate for the outgoing rotation group in March 2019 is lower than the whole sample (4.4%, compared to 5.4%). The participation rate for the outgoing rotation group in March 2019 is lower than the sample as a whole (65.5%, compared to 65.9%).


THE IMPORTANCE OF TREND DATA

As the gross flows and rotation group data are presented in original terms they are not directly comparable to the seasonally adjusted and trend data discussed elsewhere in the commentary, and are included to provide additional information for the original data. Since the original data are unadjusted, they have a considerable level of inherent sampling variability, which is specifically adjusted for in the trend series. The trend data provides the best measure of the underlying behaviour of the labour market and is the focus of the commentary in this publication.


ROTATION GROUP ANALYSIS FOR STATES AND TERRITORIES

In addition to analysis across the entire sample, the ABS also undertakes similar analysis for the responding sample in each state and territory each month, and highlights where there is a notable change for users to be aware of. For example, in the release of July 2018 data, on 16 August 2018, the ABS noted that “the rotation group effects in July 2018 were most pronounced in Queensland, Tasmania and the ACT.”

As for its reporting for the entire sample, where the ABS has not highlighted a notable incoming rotation group effect, any larger changes should therefore be considered to reflect a broader change across the sample.